A Worsted Fabric Expert System: Part II: An Artificial Neural Network Model for Predicting the Properties of Worsted Fabrics

This two-part series of papers reports on the development of a worsted fabric expert system that provides guidelines for engineering worsted fabrics. Part II describes a neural network model for predicting fabric properties based on fiber, yam, and fabric constructional parameters. An evaluation of the model shows very good agreement between the predicted and generally accepted trends of the effects of various fiber, yarn, and fabric parameters on several important fabric properties such as seam slip page, wrinkle performance, abrasion resistance, shear rigidity, bending rigidity and thickness. The predicted values of many of these fabric properties are also in good agreement with the experimental values.